Executive Pay and Performance: The Empirical Importance of the Participation Constraint

The low elasticity of top executive pay to performance in recent empirical work presents a puzzle since it is clear from the data that in practice both pay and performance measures have moved closely in line in recent years. This paper demonstrates that cross-section and time-series estimates of the pay performance elasticity differ significantly unless the effect of average executive pay is included. It is argued that this can be seen as the effect of the participation constraint in a principal agent model, more commonly known as "the going rate". The going rate must be paid to executives to deter them from taking up their outside option, which is the opportunity to leave the current firm and sign an incentive contract with another firm. Thus the outside option will depend on performance by other firms. Since performance is correlated between firms, this generates a larger time-series pay performance elasticity as perceived by the executive. The paper also considers the possibility that the going rate may induce incentive effects. The predictions of this argument are tested on a panel of data for large UK companies